Both epidemiologic and laboratory investigations have strongly suggested that estrogen plays an important role in the development and progression of breast cancer. Peptide hormones, insulin-like growth factors (IGFs), have potent mitogenic and anti-apoptotic effects on breast cancer cells and high levels of IGF-I are associated with an increased risk of breast cancer. In vitro and in vivo studies have demonstrated that IGFs interact synergistically with estrogen stimulating breast cancer growth. To examine the interplay between estrogen and IGFs in the etiology of breast cancer, we propose to measure plasma levels of these hormones using pre-treatment blood samples collected from a subset of breast cancer cases and controls, as part of a large population-based case-control study in Shanghai, China. The NIH funded large study (R0lCA64277) has recruited 1500 cases and 1500 healthy controls. Participants in the study have provided blood specimens, urine samples, and detailed questionnaire information regarding their demographic features, menstrual and reproductive history, medical history, family history of cancer, lifestyle features and dietary habit. For the proposed study, we will select 200 cases and 200 matched controls from the large study, 100 pairs for each of pre-and postmenopausal women. The cases and controls will be individually matched on age, day of blood collection, and menopausal status, as well as day of menstrual cycle for premenopausal women. Plasma samples of these 400 women will be measured for estradiol, estrone, estrone sulfate, testosterone, dehydroepiandrosterone sulfate (DHEA-S) IGF-I, IGF-II and IGFBP-3 using immunoassays. Associations of these molecules with breast cancer risk as well as the interaction between steroids and IGFs in association with breast cancer risk will be examined in multivariate analyses using the logistic regression model. Findings of the study will provide insights into the etiology of breast cancer and may have potential implications in breast cancer prevention. We will also be able to further confirm our findings by expanding the study to the whole study population if the results are promising.